Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2026 Oct 2.
Published in final edited form as: J Exp Med. 2026 Jun 8;223(8):e20251959. doi: 10.1084/jem.20251959

Bidirectional regulation by BACH2 coordinates systemic CXCR5− vs. CXCR5+ CD4+ T cell differentiation

Andrew R Schroeder 1,*, Fangming Zhu 1,*, Xianyou Xia 1, Hui Hu 1
PMCID: PMC13629307  NIHMSID: NIHMS2206137  PMID: 42257674

Abstract

CD4+ T cells differentiate into T follicular helper (Tfh) and non-Tfh cells, helping regulate humoral and cellular responses. Here, we show that in robust CD4+ T cell responses, all early CXCR5+CD4+ T cells expressed CD25, and STAT5, which inhibits Tfh cell differentiation, was crucial for both non-Tfh and Tfh responses. We found that after T cell activation, the transient increase in BACH2 protein levels temporarily suppressed STAT5-mediated Blimp1 induction, licensing early CXCR5+CD25+CD4+ T cells to proliferate to STAT5 signaling without being suppressed for the subsequent Tfh cell differentiation. Furthermore, depending on the immunization route, location, and antigen dose, BACH2 exerted bidirectional regulation on the differentiation of CXCR5− versus CXCR5+ CD4+ T cells throughout the body. Our findings suggest BACH2 orchestrates systemic immunity, offering insights for harnessing CD4+ T cell responses in vaccines and treatments.

Introduction

Humoral and cellular immune responses play crucial roles in defending against infections and in vaccine immunization. Upon activation, naive CD4+ T cells differentiate into T follicular helper (Tfh) cells and non-Tfh cells to regulate humoral and cellular responses (Crotty, 2011; Vinuesa et al., 2005; Zhu et al., 2010). However, the complex regulation that governs early CXCR5− versus CXCR5+ CD4+ T cell differentiation (Schroeder et al., 2021), as well as the mechanisms coordinating systemic immune responses throughout the body, is not fully understood.

Naïve antigen-specific CD4+ T cells, initially present at low frequencies, undergo activation and rapid expansion upon antigen stimulation (Jenkins and Moon, 2012), and STAT5 signaling is critical for the metabolism and proliferation of these newly activated CD4+ T cells (Malek and Castro, 2010; Moriggl et al., 1999; Villarino et al., 2022). In Tfh cell development, transcription factors Bcl6 and Blimp1 antagonize each other, with Bcl6 promoting and Blimp1 inhibiting Tfh cell differentiation (Johnston et al., 2009; Nurieva et al., 2009; Yu et al., 2009). STAT5 signaling has been shown to induce Blimp1 expression and suppress Tfh cell differentiation (Ballesteros-Tato et al., 2012; Johnston et al., 2012). As studies have shown that CXCR5+CD4+ T cells do not express CD25—thereby escaping the CD25/STAT5/Blimp1 pathway–mediated suppression of Tfh cell development (DiToro et al., 2018; Johnston et al., 2012; Pepper et al., 2011)—the difference in CD25 expression has been posited to be the early mechanism for Tfh/non-Tfh separation. An intriguing dilemma, however, arises. While the function of STAT5 in regulating newly activated CD4+ T cell metabolism and proliferation is to maximize the overall CD4+ T cell response—encompassing both the humoral and cellular arms—the STAT5/Blimp1 axis would inhibit Tfh cell differentiation, thereby reducing the humoral arm of the CD4+ T cell response (Crotty, 2014; Victora and Nussenzweig, 2012). Moreover, while Blimp1 was found to be directly repressed by BACH2 (Ochiai et al., 2006; Ochiai et al., 2008; Tsukumo et al., 2013), perplexingly, BACH2 has been shown to negatively regulate Tfh cell differentiation (Geng et al., 2019; Lahmann et al., 2019; Zhang et al., 2019). How STAT5, Blimp1, and BACH2 interact during CXCR5− versus CXCR5+ CD4+ T cell differentiation and regulate Tfh and non-Tfh cell responses remains unclear.

In local infection or vaccine immunization, antigens can spread to other parts of the body via different routes. Presumably, the availability of large amounts of antigen in the draining lymph nodes (LNs) would favor and sustain robust germinal center (GC) responses and antibody affinity maturation, and a strong Tfh cell response would be preferable. In contrast, in non-draining LNs, the low antigen levels would not be ideal for GC responses, and a strong Tfh cell response may be unnecessary or even counterproductive. From the perspective of systemic humoral and cellular responses across the body, it is thus intriguing to ponder whether a CD4+ T cell–intrinsic mechanism might adapt to these two situations and differentially regulate Tfh cell differentiation in draining versus non-draining LNs.

In the present study, we unexpectedly found that STAT5 signaling was crucial for both robust non-Tfh and Tfh cell responses. Furthermore, we found that depending on immunization route, location, and antigen dose, BACH2 exerts bidirectional regulation on the differentiation of CXCR5− versus CXCR5+ CD4+ T cells, orchestrating systemic immunity throughout the body.

Results and discussion

CD25/STAT5 signaling is crucial for robust non-Tfh and Tfh cell responses

Using a model of OT-II adoptive transfer and intranasal (i.n.) immunization with NP-OVA plus LPS (i.n. NP-OVA/LPS), we found that most donor cells in the draining mediastinal LN (D-medLN) were activated and expressed CD69 by day 2 (Fig. 1 A), at which time CXCR5 expression in donor cells also began to be induced (Fig. 1 B). Surprisingly, while CD25 was induced in most of the activated donor cells on day 1 and maintained until day 2 (Fig. 1, B and C), the early CXCR5+CD4+ T cells (day 2) were all CD25+ (Fig. 1 B). Using the IL-2-eGFP reporter mice (DiToro et al., 2018), we found high levels of IL-2 expression in the CD25+ donor cells (Fig. 1 C). Similarly, using the OT-II adoptive transfer/PR8-OVA (OT-II/PR8-OVA) influenza virus infection model (Zhu et al., 2023), we found that on day 3, the majority of the donor cells in the D-medLN expressed CD25 and IL-2, and the earliest wave of CXCR5+ donor cells were all CD25+ before the appearance of a CD25 versus CXCR5 dichotomy (Fig. 1 D). These results suggest that in vivo in the i.n. immunization and influenza infection models, the large majority of the early stage activated CD4+ T cells in the draining LNs co-express CD25 and IL-2, and CXCR5+CD4+ T cells are derived from the newly activated CD4+ T cells expressing CD25.

Figure 1. Induction of CD25, IL-2, and CXCR5 in early activated CD4+ T cells and the crucial role of STAT5 in both non-Tfh and Tfh responses.

Figure 1.

Open in a new tab

(A–C) Purified OT-II cells from OT-IITg or OT-IITgIL-2-eGFPTg (OT-II-IL-2-eGFP) mice were transferred into CD45.1+ SMARTA recipient mice, followed by i.n. immunization with NP-OVA plus LPS. Naïve (CD44lo) and activated donor cells in the D-medLN were analyzed at the indicated time points for: (A) CD69 staining, (B) PD-1 and CXCR5 staining (upper panel) and CD25 and CXCR5 staining (lower panel), and (C) CD25 and IL-2-eGFP staining. (D) Purified OT-II-IL-2-eGFP cells were transferred into CD45.1+ SMARTA recipient mice, followed by i.n. infection with PR8-OVA. Naïve (CD44lo) and activated donor cells in the D-medLN were analyzed at the indicated time points for: PD-1 and CXCR5 staining (top panel), CD25 and IL-2-eGFP staining (middle panel), and CD25 and CXCR5 staining (bottom panel). (E and F) CD45.1+CD45.2+ OT-II-WT and CD45.2+ OT-II-Stat5-KO cells were co-transferred into CD45.1+ SMARTA recipient mice, followed by i.n. immunization with NP-OVA plus LPS. (E) Donor OT-II-Stat5-KO to OT-II-WT cell ratios prior to transfer (naïve) and in the D-medLN at day 3 post immunization (p.i.) (n = 6). (F) Donor cells in the D-medLN at day 3 p.i. were analyzed for PD-1 and CXCR5 staining (day 3, n = 12). Data in A–F are representative (or pooled) results of at least two independent experiments. Bars represent average ± SD. The P values were determined by a one sample Wilcoxon test (E) or a two-tailed paired t test (F). *P < 0.05, **P < 0.01, ***P < 0.001; differences with P < 0.05 were considered statistically significant.

To determine how CD25/STAT5 signaling may affect the CXCR5+CD4+ T cell response, we generated OT-IITgStat5f/fCD4-CreTg (OT-II-Stat5-KO, deletion of both Stat5a and Stat5b [Cui et al., 2004]) mice. In the i.n. NP-OVA/LPS model, OT-II cells from OT-IITgCD4-CreTg (OT-II-WT) and OT-II-Stat5-KO mice were co-transferred. Compared with the pre-transfer ratio, the OT-II-Stat5-KO to OT-II-WT cell ratios in the D-medLN of recipient mice decreased by day 3 (Fig. 1 E). Although STAT5 loss increased CXCR5+ CD4+ T cell differentiation by day 3 (Fig. 1 F), its early deletion significantly reduced overall CD4+ T cell numbers, with both CXCR5− and CXCR5+ OT-II-Stat5-KO cells showing lower counts than OT-II-WT cells (Fig. 1 F). Thus, our results suggest that without STAT5, the magnitude of both CXCR5− and CXCR5+ CD4+ T cell responses is reduced, and STAT5 signaling is crucial not only for non-Tfh cells but also for robust Tfh cell responses.

The results immediately raised the question of how the early stage CXCR5+CD25+CD4+ T cells expand in response to STAT5 signaling without being suppressed for the subsequent Tfh cell differentiation.

BACH2 blocks STAT5-mediated Blimp1 induction and promotes Tfh cell differentiation in intranasal immunization and infection

Using the Blimp1-YFPTg reporter mice (Fooksman et al., 2010), we confirmed that the Blimp1 induction in activated CD4+ T cells was STAT5-dependent (Fig. S1 A). In vitro, STAT5 was activated at 24 h and peaked at day 3 (Fig. 2 A), but there was little/no expression of Blimp1 until after day 3 (Fig. 2 B). Unexpectedly and in contrast to the mRNA expression (Fig. S1B), we found that BACH2 protein levels increased at 24 h and peaked at 48 h (Fig. 2 C). Interestingly, without BACH2, the Blimp1 induction occurred by 24 h and increased to high levels by 48 h (Fig. 2 B). Similarly, in vivo, Blimp1 was almost undetectable in OT-II-WT donor cells in the first 2 days (Fig. 2 D), but in donor cells from OT-IITgBach2f/fCD4-CreTgBlimp1-YFPTg (OT-II-Bach2-KO-Blimp1-YFP) mice, Blimp1 was already up on day 1 and had strong expression on day 2 (Fig. 2 D). Thus, both in vitro and in vivo results demonstrate that STAT5-mediated Blimp1 induction in activated CD4+ T cells is arbitrated by the BACH2 protein levels.

Figure 2. Transiently increased BACH2 protein expression inhibits STAT5-mediated Blimp1 induction and promotes early Tfh cell differentiation.

Figure 2.

Open in a new tab

(A) Naïve CD4+ T cells from C57BL/6 mice were activated in vitro with anti-CD3 and anti-CD28 antibodies, and pSTAT5 was examined by intracellular staining at the indicated time points (n = 4). (B) OT-II-WT-Blimp1-YFP and OT-II-Bach2-KO-Blimp1-YFP cells were activated as in A, and Blimp1-YFP expression was examined by staining at the indicated time points (n = 3). (C) Naïve CD4+ T cells from C57BL/6 mice were activated as in A, and BACH2 protein levels were examined at the indicated time points by immunoblotting. (D) CD45.1+CD45.2+ OT-II-WT-Blimp1-YFP and CD45.2+ OT-II-Bach2-KO-Blimp1-YFP cells were transferred into CD45.1+ SMARTA recipient mice, followed by i.n. immunization with NP-OVA plus LPS. Donor cells in the D-medLN at days 1 and 2 p.i. were analyzed for Blimp1-YFP expression (day 1, OT-II-WT-Blimp1-YFP, n = 8; day 2, OT-II-WT-Blimp1-YFP, n = 6; day 1, OT-II-Bach2-KO-Blimp1-YFP, n = 5; day 2, OT-II-Bach2-KO-Blimp1-YFP, n = 4). (E) CD45.1+CD45.2+ OT-II-WT and CD45.2+ OT-II-Bach2-KO cells were co-transferred (upper panel, n = 12) or separately transferred (lower panel, n = 6) into CD45.1+ SMARTA recipient mice, followed by i.n. immunization with NP-OVA plus LPS. Donor cells in the D-medLN at day 3 p.i. were analyzed for PD-1 and CXCR5 staining. (F) CD45.1+CD45.2+ OT-II-WT, CD45.2+ OT-II-Bach2-KO-YFP (from OT-IITgBach2f/fCD4-CreTgRosaYFP mice), and CD45.2+ OT-II-Bach2/Blimp1-DKO cells were triple co-transferred into CD45.1+ SMARTA recipient mice, followed by i.n. immunization with NP-OVA plus LPS. Donor cells in the D-medLN at day 3 p.i. were analyzed for PD-1 and CXCR5 staining (n = 3). (G) CD45.1+CD45.2+ OT-II-WT, CD45.2+ OT-II-Bach2-KO-YFP, and CD45.2+ OT-II-Bach2/Stat5-DKO cells were triple co-transferred into CD45.1+ SMARTA recipient mice followed by i.n. immunization with NP-OVA plus LPS. Donor cells in the D-medLN at day 3 p.i. were analyzed for PD-1 and CXCR5 staining (n = 7). Data in A–G are representative (or pooled) results of at least two independent experiments. Bars represent average ± SD. The P values were determined by a two-way ANOVA (A, B, and D), a two-tailed paired t test (upper panel in E) or unpaired t test (lower panel in E), or a one-way ANOVA (F and G). *P < 0.05, **P < 0.01, ***P < 0.001; differences with P < 0.05 were considered statistically significant. Source data are available for this figure: SourceData F2.

Previous studies including our own work using an intraperitoneal (i.p.) immunization model have shown that BACH2 negatively regulates Tfh cell differentiation (Geng et al., 2019; Lahmann et al., 2019; Zhang et al., 2019). However, the quick and strong Blimp1 induction in the absence of Bach2 in i.n. immunization (Fig. 2 D) prompted us to re-examine the BACH2 function in Tfh cell differentiation. When OT-II cells from OT-II-WT mice and OT-IITgBach2f/fCD4-CreTg (OT-II-Bach2-KO) mice were co-transferred in the i.n. NP-OVA/LPS model, we found that in the D-medLN, OT-II-Bach2-KO cells had a lower percentage of Tfh cells than did the OT-II-WT cells (Fig. 2 E, upper panel). This reduced Tfh cell differentiation in the absence of Bach2 was also observed when OT-II-WT and OT-II-Bach2-KO cells were transferred separately (Fig. 2 E, lower panel). When B1–8i B cells were co-transferred with OT-II cells, by gating on NP-specific (NP30+) donor B cells, we found that compared with the OT-II-WT control group, the transfer of OT-II-Bach2-KO cells led to the reduced generation of NP30+GL-7+ and NP30+GL-7+IgD− B cells, indicating overall reduced GC B cell responses (Fig. S1, C and D).

Consistently, in both the PR8/polyclonal T cell infection model and the OT-II/PR8-OVA infection model, via i.n. infection, the deletion of Bach2 in CD4+ T cells led to reduced Tfh cell differentiation in the D-medLN (Fig. S1, E and F).

To evaluate the role of Blimp1 in the reduced Tfh cell differentiation of Bach2-deficient CD4+ T cells, we generated the OT-IITgBach2f/fBlimp1f/fCD4-CreTg (OT-II-Bach2/Blimp1-DKO) mice and did the triple co-transfer of WT, Bach2-KO, and Bach2/Blimp1-DKO OT-II cells in the i.n. NP-OVA/LPS model. On day 3, whereas the Tfh cell differentiation in the D-medLN was reduced in OT-II-Bach2-KO cells, the deletion of Blimp1 in Bach2-deficient donor cells rescued the CXCR5+CD4+ T cell differentiation (Fig. 2 F).

The early responding OT-II-WT cells in the spleen in i.p. immunization gave rise to CXCR5+CD4+ T cells, but they expressed little CD25 (or IL-2) (Fig. S1 G). In contrast, the early responding OT-II-WT cells in the D-medLN in i.n. immunization expressed high levels of CD25 (Fig. 1 B and Fig. S1 H), and the OT-II-Bach2-KO cells showed higher percentages as well as mean fluorescence intensity of CD25 expression (Fig. S1 H). A recent study has reported that BACH2 regulates CD25 expression in regulatory T (Treg) cells (Zhang et al., 2021). In conventional naive CD4+ T cells from CD4-CreTg mice and Bach2f/fCD4-CreTg (Bach2-KO) mice, together with the retroviral overexpression approach, we found that both loss-of-function and gain-of-function experiments demonstrated that BACH2 was a negative regulator of CD25 in non-Treg CD4+ T cells as well (Fig. S1, I and J). To determine whether the suppressed Tfh cell differentiation of Bach2-deficient CD4+ T cells in i.n. immunization was due to the enhanced CD25/STAT5 signaling, we generated the OT-IITgBach2f/fStat5f/fCD4-CreTg (OT-II-Bach2/Stat5-DKO) mice and found that the deletion of Stat5 rescued the reduced Tfh cell differentiation of Bach2-deficient donor cells (Fig. 2 G). Recently, we showed that STAT3 modulates BACH2 protein levels (Schroeder et al., 2023). However, in the i.n. protein immunization model, we excluded the involvement of STAT3 in the early CXCR5− versus CXCR5+ CD4+ T cell differentiation for WT or Bach2-deficient OT-II cells (Fig. S1, K and L).

Taken together, these results suggest that opposite to its function in suppressing Tfh cell differentiation in i.p. immunization (in which low or no CD25/IL-2 was induced [Fig. S1 G]) (Geng et al., 2019), BACH2 promotes Tfh cell differentiation in i.n. immunization and infection, in which high levels of CD25 and IL-2 are induced.

The bidirectional regulation of Tfh versus non-Tfh by BACH2 throughout the body

In vitro, using a Cxcr5 promoter reporter (Geng et al., 2019; Leong et al., 2016), we found that stronger TCR stimulation induced higher Cxcr5 promoter activities (Fig. S2 A). As TCR stimulation also led to increased BACH2 protein levels (Fig. 2 C), it became intriguing how the strength of T cell stimulation interplays with BACH2 to regulate CXCR5− versus CXCR5+ CD4+ T cell differentiation.

In the OT-II/i.n. NP-OVA/LPS model, we found that the spread of the antigen also activated the donor cells in the non-draining mesenteric LN (ND-mesLN) (Fig. 3 A). Presumably due to higher levels of antigen and therefore stronger T cell stimulation in the D-medLN, the expression levels of CD25 and IL-2 were higher in the donor cells in the D-medLN than in the ND-mesLN (Fig. 3 A); consistent with the function of STAT5 signaling in CD4+ T cell proliferation (Moriggl et al., 1999), the magnitude of the CD4+ T cell response was substantially higher in the D-medLN than in the ND-mesLN (Fig. 3 B). Interestingly, after i.n. immunization with NP-OVA plus LPS, we found that the Bach2 deficiency resulted in reduced CXCR5+CD4+ T cell differentiation in the D-medLN but increased CXCR5+CD4+ T cell differentiation in the ND-mesLN (Fig. 3 C).

Figure 3. Bidirectional regulation of Tfh versus non-Tfh cell differentiation by BACH2 in draining versus non-draining LNs or under conditions of different antigen doses.

Figure 3.

Open in a new tab

(A and B) OT-II-IL-2-eGFP cells were transferred into CD45.1+ SMARTA recipient mice, followed by i.n. immunization with NP-OVA plus LPS. (A) At day 1 p.i., donor OT-II cells in the D-medLN and the ND-mesLN were analyzed for CD4, CD69, CD25, and IL-2-eGFP staining (n = 12). (B) Gated on total CD4+ T cells, donor cell recovery in the D-medLN and the ND-mesLN were analyzed at the indicated time points (day 1, n = 13; day 2, n = 7; day 2.5, n = 9; day 3, n = 13; day 4, n = 12). (C) CD45.1+CD45.2+ OT-II-WT and CD45.2+ OT-II-Bach2-KO cells were co-transferred into CD45.1+ SMARTA recipient mice, followed by i.n. immunization with NP-OVA plus LPS. Donor cells in the D-medLN and the ND-mesLN at day 3 p.i. were analyzed for CD25 and CXCR5 staining (n = 7). (D) Diagram of draining and non-draining LNs in the s.c. immunization model system. (E) OT-II-IL-2-eGFP cells were transferred into CD45.1+ SMARTA recipient mice, followed by s.c. immunization with NP-OVA plus LPS. Donor cells in the draining and non-draining LNs at day 2 p.i. were analyzed for CD25 and IL-2-eGFP staining (n = 7). (F) CD45.1+CD45.2+ OT-II-WT and CD45.2+ OT-II-Bach2-KO cells were co-transferred into CD45.1+ SMARTA recipient mice, followed by s.c. immunization with NP-OVA plus LPS. Donor cells in the draining and non-draining LNs at day 3 p.i. were analyzed for CD25 and CXCR5 staining (D-ingLN, n = 4; ND-ingLN, n = 4; ND-medLN, n = 7; ND-mesLN, n = 7). (G) OT-II-IL-2-eGFP cells were transferred into CD45.1+ SMARTA recipient mice, followed by s.c. immunization with indicated dose (100 μg, n = 8; 1 μg, n = 6) of NP-OVA plus LPS. Donor cells in the D-ingLN at day 1 p.i. were analyzed for CD25 and IL-2-eGFP staining. (H) CD45.1+CD45.2+ OT-II-WT and CD45.2+ OT-II-Bach2-KO cells were co-transferred into CD45.1+ SMARTA recipient mice, followed by s.c. immunization with indicated doses of NP-OVA plus LPS. Donor cells in the D-ingLN at day 3 p.i. were analyzed for PD-1 and CXCR5 staining (n = 9). Data in A–C and E–H are representative (or pooled) results of at least two independent experiments. Bars represent average ± SD. The P values were determined by a two-tailed unpaired t test (A and G), a two-way ANOVA (B, C, and F), a one-way ANOVA (E), or a two-tailed paired t test (H). **P < 0.01, ***P < 0.001; differences with P < 0.05 were considered statistically significant.

To validate this bidirectional function of BACH2 in regulating CXCR5+CD4+ T cell differentiation in draining versus non-draining LNs, we next tested the subcutaneous (s.c.) immunization model system, in which the inguinal LN (ingLN) ipsilateral to the immunization would serve as the draining LN, whereas the contralateral ingLN—as well as the medLN and mesLN—would serve as non-draining LNs (Fig. 3 D). We found that whereas the donor cells in the D-ingLN had high levels of CD25 and IL-2 expression, the donor cells in the non-draining LNs expressed low/no CD25 or IL-2 (Fig. 3 E). Importantly, compared with OT-II-WT cells, OT-II-Bach2-KO cells had significantly reduced Tfh cell differentiation in the D-ingLN, but increased Tfh cell differentiation in the ND-ingLN, ND-medLN, and ND-mesLN (Fig. 3 F). Accordingly, in the D-ingLN, the generation of endogenous B220+GL-7+Fas+ GC B cells was also significantly reduced in mice receiving OT-II-Bach2-KO cells compared with the OT-II-WT control group (Fig. S2 B). In the ND-ingLN, however, whereas the CXCR5+ OT-II cells did exhibit higher Bcl6 expression levels (Fig. S2 C), the low levels of antigens there were insufficient to drive GC B cell responses above the background control (Fig. S2 B). Conversely, these results emphasize that BACH2 favors the differentiation of CXCR5− non-Tfh cells within the non-draining LNs, thereby supporting the bidirectional regulation by BACH2 at the systemic level.

To test directly the idea that the bidirectional function of BACH2 could be attributed to antigen dose, we first focused on the CD4+ T cell responses in the D-ingLN in s.c. immunization model. In the D-ingLN, the OT-II-WT cells expressed high levels of CD25 and IL-2 in response to a high dose of 100 μg NP-OVA plus LPS compared with low/no CD25/IL-2 expression under the condition of a low dose of 1 μg NP-OVA plus LPS (Fig. 3 G). Accordingly, we found that compared with OT-II-WT cells, OT-II-Bach2-KO cells had reduced Tfh cell differentiation with the high-dose immunization and increased Tfh cell differentiation with the low-dose immunization (Fig. 3 H). Similar results of reduced and increased Tfh cell differentiation of OT-II-Bach2-KO cells were obtained in the D-medLN in the i.n. immunization with high dose of 20 μg and low dose of 0.2 μg NP-OVA plus LPS (Fig. S2, D and E), respectively. In between doses of 20 and 0.2 μg NP-OVA plus LPS, if we chose a dose of 1 μg NP-OVA plus LPS, a delicate equilibrium was reached, such that no difference in Tfh cell differentiation was found between OT-II-WT and OT-II-Bach2-KO cells (Fig. S2 F). To reduce the initial frequency of antigen-specific CD4+ T cells to a level more closely approximating physiological conditions, we also performed transfer experiments using 5,000 OT-II cells, assuming a 10% seeding efficiency prior to immunization (Hu et al., 2001). Under these conditions, Bach2-deletion still led the CD4+ T cell response to decreased CXCR5+CD4+ T cell differentiation following high-dose immunization, yet increased CXCR5+CD4+ T cell differentiation following low-dose immunization (Fig. S2 G).

To further understand how BACH2 protein levels are regulated by TCR stimulation (Sidwell et al., 2020), we stimulated naive CD4+ T cells with different amounts of anti-CD3/CD28 antibodies in vitro. We found that while progressively increasing TCR stimulation led to higher IL-2 expression levels (Fig. S2 H), the increase in BACH2 protein levels was inversely correlated with the strength of TCR stimulation (Fig. S2 I). It is worth noting that the relatively lower BACH2 protein levels under strong TCR stimulation were sufficient to suppress the Blimp1 expression (Fig. 2 B). BACH2 has been suggested to suppress Bcl6 expression (Lahmann et al., 2019), although this was not observed in other studies (Zhang et al., 2019). Previously, we have shown that BACH2 directly and negatively regulates CXCR5 expression (Geng et al., 2019). In the current study, we found that BACH2 also negatively regulated the negative regulators of CXCR5, namely Blimp1 and CD25 (Fig. 2, B and D; and Fig. S1, I and J). Thus, the TCR stimulation, the transiently increased BACH2 protein levels, and the Bach2 network exert a complex, bidirectional regulation on CXCR5− versus CXCR5+ CD4+ T cell differentiation in response to the differential antigen distribution and stimulation.

BACH2 inhibits STAT5-mediated induction of Blimp1 via histone modification

To understand the interplay between BACH2 and STAT5 during the early stage of CD4+ T cell response, we cross-examined the published BACH2 and STAT5 ChIP-seq datasets (Grant et al., 2020; Villarino et al., 2022) and found that BACH2 did not appear to globally regulate STAT5 targets (Fig. 4 A). In the Prdm1 locus, we found multiple BACH2 and STAT5 co-binding regions (Fig. 4 B). The ATAC-seq analysis of Bach2- and Stat5-deficient CD4+ T cells (Fig. 4 C) showed that neither BACH2 nor STAT5 had much impact on the chromatin accessibility of the Prdm1 locus (Fig. 4 D).

Figure 4. BACH2 binds to the Prdm1 locus to inhibit permissive histone acetylation.

Figure 4.

Open in a new tab

(A) Global heatmap of BACH2- and STAT5-binding peaks in BACH2 ChIP-seq (GSE128176) and STAT5 ChIP-seq (GSE207265) datasets. (B) IGV tracks showing BACH2- and STAT5-binding peaks in the Prdm1 locus. (C and D) Naïve WT, Bach2-KO, Stat5-KO, and Bach2/Stat5-DKO CD4+ T cells were activated in vitro with anti-CD3 and anti-CD28 antibodies. ATAC-seq analysis of the activated CD4+ T cells was performed at 48 h. (C) Global heatmap showing the ATAC-seq peaks. (D) IGV tracks showing the chromatin accessibility peaks of the Prdm1 locus. (E) OT-II-WT, OT-II-Stat5-KO, and OT-II-Bach2-KO cells were activated in vitro with anti-CD3 and anti-CD28 antibodies. H3K27Ac ChIP-qPCR analysis was performed to assess the H3K27ac levels of the Prdm1 locus at 48 h (n = 3). (F) OT-II-WT-Blimp1-YFP and OT-II-Bach2-KO-Blimp1-YFP cells were activated in vitro with anti-CD3 and anti-CD28 antibodies and treated with DMSO or p300 inhibitor RGFP966 (p300i). Blimp1-YFP expression in activated CD4+ T cells were analyzed at 48 h (n = 4). Data in E and F are representative results of at least two independent experiments. Bars represent average ± SD. The P values were determined by a two-way ANOVA (E and F). ATAC-seq samples are independent biological triplicates. **P < 0.01, ***P < 0.001; differences with P < 0.05 were considered statistically significant.

BACH2 has been shown to interact with HDAC3 to regulate targets via histone modification (Tanaka et al., 2016). Using histone-ChIP, we found that levels of the permissive modification of H3 lysine 27 acetylation (H3K27ac) at the BACH2 and STAT5 co-binding regions in the Prdm1 locus were not changed in the absence of Stat5, but increased in the absence of Bach2 (Fig. 4 E). When we used p300 inhibitor RGFP966 (p300i), the inhibition of H3K27 acetylation in activated OT-II-Bach2-KO-Blimp1-YFP cells reduced the Blimp1 expression levels (Fig. 4 F), suggesting that BACH2 imposes its negative regulation of the Prdm1 locus by suppressing the H3K27ac levels.

The regulation of Tfh and non-Tfh cell differentiation plays a critical role in adaptive immunity. Here, we uncovered fundamental mechanisms underlying CD4+ T cell responses. At the systemic level, we have revealed a complex relationship between the early CXCR5− versus CXCR5+ cell development and the regulation of the magnitude of CD4+ T cell responses, both mediated and interconnected through the Bach2 network in a CD4+ T cell–intrinsic manner. In strong CD4+ T cell responses, where high levels of CD25 and IL-2 expression and a higher magnitude of the CD4+ T cell response were induced, we found that all CXCR5+CD4+ T cells were derived from the newly activated CD4+ T cells expressing CD25. Most importantly, we found that the BACH2 protein levels were transiently increased after T cell activation, and BACH2 temporarily inhibited the CD25/STAT5-mediated Blimp1 induction, which allowed the emerging CXCR5+CD25+CD4+ T cells to escape suppression and proceed with subsequent Tfh cell differentiation. This explains how the early activated CXCR5+CD25+CD4+ T cells are granted a “time window” to proliferate and expand to STAT5 signaling, highlighting a fundamental mechanism underlying the crucial role of STAT5 signaling in eliciting a robust Tfh cell response. Under such conditions, BACH2 helped promote the Tfh cell differentiation and response. When the antigen dose or T cell stimulation was low/weak, resulting in low or no CD25/IL-2 induction and a reduced magnitude of the CD4+ T cell response, presumably higher BACH2 protein levels favored the non-Tfh cell differentiation by directly suppressing CXCR5 expression, thereby restraining Tfh cell differentiation. Although technical limitations have precluded the direct measurement of BACH2 protein levels in vivo, this BACH2-mediated bidirectional regulation was confirmed in various experimental conditions: (1) The same protein immunization comparing i.n. versus i.p. routes; (2) the same animal comparing draining versus non-draining LNs; and (3) the same draining LN with high versus low antigen doses; and our results indicate universal applicability.

Models of TCR affinity (Fazilleau et al., 2009), TCR-peptide/MHC II dwelling time (Tubo et al., 2013), and divergent CD25 versus IL-2 expression pattern (DiToro et al., 2018) have been proposed to explain the early CXCR5− versus CXCR5+ CD4+ T cell differentiation. Our findings, particularly that STAT5 signaling is crucial for not only non-Tfh but also Tfh cell responses, lead us to suggest a shifted conceptual framework: (1) Overall CD4+ T cell responses can be categorized into two types: a strong CD4+ T cell response with high levels of CD25 and IL-2, and a weak CD4+ T cell response with low or no CD25/IL-2 expression. The CD25/IL-2 induction and STAT5 signaling determine the magnitude of the CD4+ T cell response that encompasses both non-Tfh cells and Tfh cells. (2) The differentiation of CXCR5− versus CXCR5+ CD4+ T cells is bidirectionally regulated by the Bach2 network. Mechanistically, BACH2 negatively regulates CXCR5 expression while simultaneously suppressing the negative regulators (CD25 and Blimp1) of CXCR5, constituting a novel type 2 incoherent feedforward loop that fine-tunes CXCR5− versus CXCR5+ divergence in the early activated CD4+ T cells under different conditions (Fig. S3 A). The regulation of Cxcr5 promoter activity by TCR signaling and the inverse correlation between TCR stimulation strength and BACH2 protein levels are also part of this CD4+ T cell–intrinsic regulatory mechanism (Fig. S3 B), as well as the factors that may potentially regulate BACH2 protein levels. These results position Bach2 at the center of the network, orchestrating diverse components and pathways to modulate CXCR5− versus CXCR5+ CD4+ T cell differentiation. Together with the results that BACH2 “filters” STAT5 signaling to support the initial expansion of newly activated CD4+ T cells, the Bach2 network represents a critical “Step 2,” following “Step 1” of TCR activation and co-stimulation, in defining the fundamental principles underlying the CD4+ T cell response.

It is worth noting that a delicate equilibrium incorporating the signals of this network was demonstrated by no difference in CXCR5+CD4+ T cell differentiation between WT and Bach2-deficient CD4+ T cells at an intermediate antigen dose in the i.n. immunization model.

The results of STAT5 signaling helping Tfh cell response are consistent with the STAT5 function in promoting activated CD4+ T cell metabolism and proliferation (Moriggl et al., 1999; Villarino et al., 2022). In strong CD4+ T cell responses, within the CD25+CD4+ T cell population, the CXCR5+ cells were the first to downregulate CD25, leading to the subsequent dichotomous staining pattern of CD25 versus CXCR5. At the T-B border, an early study has shown that soluble CD25 secreted by dendritic cells creates an IL-2 low zone (Li et al., 2016). Thus, it is likely that when CXCR5+CD25+CD4+ T cells, but not CXCR5−CD25+CD4+ T cells, migrate toward the B cell follicle, the T-B border microenvironment helps “starve” the CXCR5+CD25+CD4+ T cells of IL-2/CD25 signaling, resulting in their quick downregulation of CD25 and escape from the STAT5 signaling allowing the progression toward the GC Tfh cell differentiation. On the other hand, the CXCR5−CD25+CD4+ T cells receive IL-2/CD25 signaling and commit to the non-Tfh cell pathway.

BACH2-mediated bidirectional regulation on Tfh versus non-Tfh cell differentiation may also help understand the paradoxical clinical presentation of patients with Bach2 deficiency (Afzali et al., 2017; Igarashi et al., 2017): While the autoimmune conditions with autoantibodies in Bach2-deficient patients may result from the absence of BACH2-mediated suppression of Tfh responses to weakly stimulating autoantigens, their inadequate humoral responses to many pathogens or vaccinations likely stem from a lack of BACH2 function in promoting Tfh cell differentiation under strong antigen stimulation.

In summary, our findings reveal the fundamental regulatory mechanism by which newly activated CD4+ T cells expand and choose between CXCR5− and CXCR5+ fates under varying conditions. The study provides valuable insights for developing new strategies to harness CD4+ T cell responses in vaccination and the treatment of infectious and autoimmune diseases.

Materials and methods

Mice

CD45.2+ C57BL/6 (Strain #: 000664), CD45.1+ C57BL/6 (Strain #: 002014), OT-II (Strain #: 004194), CD45.1+ SMARTA (Strain #: 030450), RosaYFP (Strain #: 006148), CD4-CreTg (Strain #: 017336), Blimp1-YFPTg (Strain #: 008828), and Blimp1f/f mice (Strain #: 008100) were purchased from the Jackson Laboratory. OT-II mice were bred to CD45.1+ C57BL/6 congenic mice to generate CD45.1+CD45.2+ OT-IITg mice. CD45.1+CD45.2+ OT-IITg mice were bred to CD4-CreTg and Blimp1-YFPTg mice to generate CD45.1+CD45.2+ OT-IITgCD4-CreTg and CD45.1+CD45.2+ OT-IITgBlimp1-YFPTg mice, respectively. IL-2-eGFPTg mice were obtained from Dr. Casey T. Weaver and bred to OT-II mice to generate OT-IITgIL-2-eGFPTg mice. Blimp1f/f mice were bred to OT-II and CD4-CreTg mice to generate OT-IITgBlimp1f/fCD4-CreTg mice. Stat5f/f mice were obtained from Dr. John J. O’Shea and bred to OT-II and CD4-CreTg mice to generate OT-IITgStat5f/fCD4-CreTg mice in which both Stat5a and Stat5b genes are deleted. OT-IITgStat5f/fCD4-CreTg mice were bred to Blimp1-YFPTg mice to generate OT-IITgStat5f/fCD4-CreTgBlimp1-YFPTg mice. Bach2f/f mice were generated as described (Geng et al., 2019) and bred to OT-II and CD4-CreTg mice to generate OT-IITgBach2f/fCD4-CreTg mice. OT-IITgBach2f/fCD4-CreTg mice were bred to Blimp1-YFPTg, RosaYFP, Stat5f/f, and Blimp1f/f mice to generate OT-IITgBach2f/fCD4-CreTgBlimp1-YFPTg, OT-IITgBach2f/fCD4-CreTgRosaYFP, OT-IITgBach2f/fStat5f/fCD4-CreTg, and OT-IITgBach2f/fBlimp1f/fCD4-CreTg mice, respectively. Experiments were conducted using age- and sex-matched male and female mice at 6–12 wk of age. All animals were maintained in specific pathogen–free barrier facilities and were used in accordance with protocols approved by the Institutional Animal Care and Use Committee of the University of Alabama at Birmingham (Birmingham, Alabama, USA). The abbreviations of mice used in the paper: WT (control in polyclonal T cell study), CD4-CreTg; Stat5-KO, Stat5f/fCD4-CreTg; Bach2-KO, Bach2f/fCD4-CreTg; OT-II-WT (control in OT-II transfer study), OT-IITgCD4-CreTg; OT-II-IL-2-eGFP, OT-IITgIL-2-eGFPTg; OT-II-Stat5-KO, OT-IITgStat5f/f CD4-CreTg; OT-II-Bach2-KO, OT-IITgBach2f/fCD4-CreTg; OT-II-Bach2/Stat5-DKO, OT-IITgBach2f/fStat5f/fCD4-CreTg; OT-II-Bach2/Blimp1-DKO, OT-IITgBach2f/fBlimp1f/fCD4-CreTg; OT-II-WT-Blimp1-YFP, OT-IITgBlimp1-YFPTg; OT-II-Stat5-KO-Blimp1-YFP, OT-IITgStat5f/f CD4-CreTgBlimp1-YFPTg; OT-II-Bach2-KO-Blimp1-YFP, OT-IITgBach2f/fCD4-CreTgBlimp1-YFPTg; OT-II-Bach2-KO-YFP, and OT-IITgBach2f/fCD4-CreTgRosaYFP.

Adoptive transfer, infection, and immunization

OT-II cells were purified from LNs and spleens using the Invitrogen Dynabeads CD4 positive isolation kit (Thermo Fisher Scientific) or mouse CD4 (L3T4) MicroBeads (Miltenyi Biotec) (>95% CD44loCD62Lhi). Unless otherwise stated, for OT-II cell transfer experiments, (1) a total 3 × 104 purified OT-II cells were transferred intravenously into SMARTA recipient mice, simultaneously, followed by influenza virus infection; or (2) a total 1 × 106 purified OT-II cells were transferred intravenously into SMARTA recipient mice, rested for 1 day, followed by NP-OVA/Alum immunization intraperitoneally or NP-OVA/LPS immunization intranasally or subcutaneously. For OT-II and B1–8i B cell co-transfer experiments, 8 × 106 purified B1–8i B cells with 4 × 105 OT-II cells were co-transferred intravenously into SMARTA recipient mice, rested for 1 day, followed by NP-OVA/LPS immunization intranasally. For Stat3-CRISPR experiments, 1.5 × 106 Stat3-CRISPR– or CD90-CRISPR–mixed OT-II-WT and OT-II-Bach2-KO cells were co-transferred intravenously into SMARTA recipient mice, rested for 1 day, followed by NP-OVA/LPS immunization intranasally. For all influenza virus infection experiments, mice were immobilized with isoflurane and intranasally infected with mouse-adapted influenza virus A/Puerto Rico/8/34 (PR8) at a dose of 15,000 virus forming units (v.f.u.) or PR8 virus expressing the OVA (323–339) epitope (PR8-OVA) at a dose of 1,000 v.f.u. For NP-OVA/Alum immunization experiments, mice were i.p. immunized with 100 μg NP-OVA in 25% Alum. For NP-OVA/LPS immunization experiments, mice were immobilized with isoflurane and i.n. or s.c. immunized with 100 μg NP-OVA and 3 μg LPS, unless the NP-OVA dose was otherwise specified in the figures.

Cell preparation, staining, and flow cytometry

For lymphatic organs, LNs and spleens were mashed through a 70 μm filter to obtain single-cell suspensions. Spleen samples were further incubated with ACK lysis buffer to remove red blood cells. Cells were counted using trypan blue staining, and 3–5 × 106 cells were suspended in 50 μl phosphate buffered saline (PBS) containing 2% BSA and 2 mM EDTA (FACS buffer) directly for surface staining. Nonspecific antibody binding was blocked with anti-CD16/CD32 antibodies (BioLegend) in FACS buffer for 10 min before staining. Dead cells were excluded through the use of a Live/Dead Fixable Dead Cell staining kit or Fixable Viability Dye eFluor 780 (Invitrogen). Allophycocyanin (APC) conjugated influenza nucleocapsid protein (NP311–325) tetramer was provided by the National Institutes of Health (NIH) Tetramer Core Facility, and the staining was performed at room temperature (RT) for 1 h. For surface markers staining, except for CXCR5 (which was stained at RT for 1 h), the staining was performed at 4°C for 30 min. For Bcl6 staining, cells were fixed and stained according to the standard protocol of Foxp3/Transcription Factor Staining Buffer Set. For pSTAT3 and pSTAT5 staining, cells were fixed with pre-warmed 3% formaldehyde at 37°C for 10 min, then permeabilized with ice-cold 90% MeOH on ice for 30 min and blocked with 0.5% BSA incubation buffer at RT for 10 min. The pSTAT3 or pSTAT5 antibody was prepared in 0.5% BSA incubation buffer and stained on ice for 1 h. The wash buffer for the pSTAT3 and pSTAT5 staining was 0.5% BSA incubation buffer with 1 × protease inhibitor cocktail. Samples were analyzed or sorted using custom BD LSR II, BD FACSymphony, and BD FACSAria II instruments in the UAB Comprehensive Flow Cytometry Core Facilities. Flow cytometry results were analyzed using FlowJo software (v10.8.1). For quantification, the fluorescence intensities of all acquired cells of each mouse were averaged.

CRISPR/Cas9-mediated deletion in OT-II cells

Three unique guides specific to Stat3 were obtained according to IDT’s online predesigned library (https://www.idtdna.com/site/order/designtool/index/CRISPR_PREDESIGN), specifically targeting the 5′ end of Stat3. For further details, see reference (Liu et al., 2022).

T cell stimulation, retroviral transduction, and T cell/APC co-culture

Purified OT-II or naïve CD4+ T cells were stimulated for 48 h with 0.5 μg/ml anti-CD3 (eBioscience) and 1 μg/ml anti-CD28 (eBioscience) antibodies in plates pre-coated with 20 μg/ml goat anti-hamster IgG (MP Biomedicals) in complete T cell medium under ThNone cell culture condition, and then their populations were detached and expanded for another 1–2 days in complete T cell medium containing 20 U/ml recombinant human IL-2 (rhIL-2). For the Bach2 overexpression experiments, retrovirus was produced by Plat-E cells co-transfected with retroviral vectors and helper plasmids via PEI (Polyethylenimine, Polysciences)-mediated transduction. Naïve CD4+ T cells were cultured and activated under ThNone cell culture condition and transduced at 18 and 48 h with virus-containing media supplemented with 0.6 μg/ml polybrene (Sigma-Aldrich). After the second transduction, cells were detached and expanded for another 2 days in complete T cell medium containing 20 U/ml recombinant human IL-2. For different TCR stimulation experiments, unit 1 (0.5 μg/ml anti-CD3 and 1 μg/ml anti-CD28 antibodies) was diluted at a 1:4 ratio once to obtain unit 1/4 (0.125 μg/ml anti-CD3 and 0.25 μg/ml anti-CD28 antibodies) and twice to obtain unit 1/16 (0.03125 μg/ml anti-CD3 and 0.0625 μg/ml anti-CD28 antibodies). For p300i treatment experiments, 1 μM RGFP966 (p300i) in DMSO was added to the culture medium at the beginning of cell culture, and the same volume of DMSO without p300i was added as a control. For T cell/APC co-culture experiments, Cxcr5-Promoter-Thy1.1 reporter retrovirus was produced by Plat-E cells co-transfected with retroviral vectors and helper plasmids via PEI-mediated transduction. Purified OT-II T cells were cultured and activated under ThNone cell culture conditions and transduced at 18 and 48 h with virus-containing media supplemented with 0.6 μg/ml polybrene. After the second transduction, co-culture the activated OT-II T cells with APCs (generated from splenic cells of SMARTA mice irradiated at a total dose of 3200 cGy) in the presence of 0.01 μg/ml, 0.1 μg/ml, or 1 μg/ml OVA323–339 peptide for another 2 days in complete T cell medium. ThNone cell culture condition: No cytokines, but with blocking antibodies at the first 48 h, then add 20 U/ml rhIL-2 after 48 h.

Immunoblot analysis

Cell lysis and immunoblot were performed as described (Wang et al., 2014). Primary antibodies used: rabbit anti-BACH2 (1:500 dilution, 4°C overnight) (Schroeder et al., 2023) and rabbit anti-β-Actin (1:3,000 dilution, 4°C overnight); secondary antibody: anti-rabbit IgG horseradish peroxidase-linked secondary antibody (1:6,000 dilution, RT for 1 h).

Real-time RT-PCR

Naïve CD4+ T cells from C57BL/6 mice were sorted and activated in vitro with anti-CD3 and anti-CD28 antibodies under ThNone cell culture condition. At 48 h, RNA from activated CD4+ T cells was purified, and real-time reverse-transcription was performed as described (Hu et al., 2006). Expression of Bach2 mRNA was normalized to Rpl32 expression.

Chromatin immunoprecipitation

Purified OT-II cells from OT-IITgCD4-CreTg (OT-II-WT), OT-IITgBach2f/fCD4-CreTg (OT-II-Bach2-KO), OT-IITgStat5f/fCD4-CreTg (OT-II-Stat5-KO), and OT-IITgBach2f/fBlimp1f/fCD4-CreTg (OT-II-Bach2/Blimp1-DKO) mice were activated with anti-CD3 and anti-CD28 antibodies. Cells were harvested at 48 h and sheared chromatin was generated following the manufacturer’s instructions using the truChIP Chromatin Shearing Kit (Covaris). 3% of sheared chromatin was saved for input, and 97% of sheared chromatin was used for immunoprecipitation. Chromatin immunoprecipitation of H3K27ac (Abcam) was performed as described (Wang et al., 2014). H3K27ac binding enrichment at specific genomic regions was normalized to input.

ChIP-seq

BACH2 ChIP-seq (GSE128176), STAT5 ChIP-seq (GSE207265), and Blimp1 ChIP-seq (GSE79339) datasets were obtained from GEO database repository. Data analysis was performed using the Cheaha Supercomputer at the University of Alabama at Birmingham. Low-quality reads and sequencing adapters were removed using Trim Galore! (v0.4.4). Trimmed reads were aligned to the mouse genome mm10 using Bowtie2 (v2.4.3). The SAM files were converted and sorted to BAM files using SAMtools (v1.9). PCR duplicates were marked and removed using Picard (v2.20.1) and SAMtools (v1.9), respectively. Peak calling was performed using Genrich (v0.6). The merged BAM files were generated using the MergeSamFiles tool in Picard (v2.20.1) and then converted to BW (BigWig) files using the bamCoverage tool in deepTools (v3.3.1). The read density of gene loci was visualized and generated using IGV (Integrative Genomics Viewer, v2.18.2). Further analyses were performed with R software (v4.2.2). BACH2- and STAT5-binding peaks cross analysis was performed using the intersect tool in BedTools (v2.27.1). The heatmap for scores associated with genomic regions was generated using the plotHeatmap tool in deepTools (v3.3.1).

ATAC-seq

Purified OT-II cells from OT-IITgCD4-CreTg (OT-II-WT), OT-IITgBach2f/fCD4-CreTg (OT-II-Bach2-KO), OT-IITgStat5f/fCD4-CreTg (OT-II-Stat5-KO), and OT-IITgBach2f/fStat5f/fCD4-CreTg (OT-II-Bach2/Stat5-DKO) mice were activated in vitro with anti-CD3 and anti-CD28 antibodies. 5 × 104 cells were harvested into 1.5 ml tubes, and the ATAC-seq library was prepared as described (Buenrostro et al., 2015). The sequencing was performed at GENEWIZ from Azenta Life Sciences. Data analysis was performed using the Cheaha Supercomputer at the University of Alabama at Birmingham. Low-quality reads and sequencing adapters were removed using Trim Galore! (v0.4.4). Trimmed reads were aligned to the mouse genome mm10 using Bowtie2 (v2.4.3). The SAM files were converted and sorted to BAM files using SAMtools (v1.9). PCR duplicates were marked and removed using Picard (v2.20.1) and SAMtools (v1.9), respectively. Peak calling was performed using Genrich (v0.6). The merged BAM files were generated using the MergeSamFiles tool in Picard (v2.20.1) and then converted to BW (BigWig) files using the bamCoverage tool in deepTools (v3.3.1). The read density of gene loci was visualized and generated using IGV (v2.18.2). Further analyses were performed with R software (v4.2.2). Differentially accessible regions (DARs) were identified following DESeq2 (v1.38.3) normalization using an adjusted P value <0.05. DARs were crossed with BACH2 ChIP-seq–binding peaks using the intersect tool in BedTools (v2.27.1). The heatmap for scores associated with genomic regions was generated using the plotHeatmap tool in deepTools (v3.3.1). Motif enrichment was calculated using HOMER (v4.11.1, default parameters) on peaks of ATAC-seq with both BACH2 binding and changed chromatin accessibility. The volcano plot and scatter plot were generated using ggplot2 (v3.5.1).

Statistics

Independent and paired two-tailed Student’s t tests, one-sample Wilcoxon test, independent and repeated measure one-way ANOVAs with Tukey’s multiple comparisons test, and repeated measure two-way ANOVAs with Sidak’s multiple comparisons test were performed using GraphPad Prism software (v8.2.1). The error bars represent SD. Though appropriate statistical tests were run for all comparisons made, P values were only displayed for significant results. For further details, see figure legends.

Supplementary Material

supplementary Table S1
SourceDataF2
SouceDataFS2
1

Online supplemental material

Fig. S1 provides additional data supporting the results in Fig. 2. Fig. S2 provides additional data supporting the results in Fig. 3. Fig. S3 summarizes the discoveries from this project. Table S1 lists the reagents and their sources.

Acknowledgments

We thank Vidya Sagar Hanumanthu, Harish Chandra Pal, Clarisse Iradukunda, Karlee Stortz, Haden Johnson, and Teshia Houser for flow cytometry and cell sorting; Diego O. Sialer, Giovani Zabot Pasini, Kim Nguyen, Ching-En Lee, and Aakriti Chadha for assistance with the experiments; Drs. Frances Lund and Troy Randall (University of Alabama at Birmingham, Birmingham, AL, USA) and Dr. Paul G. Thomas (Fred Hutchinson Cancer Center, Seattle, WA, USA) for PR8 and PR8-OVA influenza viruses; Dr. Matthew Pipkin (The Herbert Wertheim UF Scripps Institute, Jupiter, FL, USA) for the Plat-E cells; Dr. Casey T. Weaver (University of Alabama at Birmingham, Birmingham, AL, USA) for IL-2-eGFP mice; Dr. John J. O’Shea (NIH/National Institute of Arthritis and Musculoskeletal and Skin Diseases, Bethesda, MD, USA) for Stat5f/f mice; Dr. Di Yu (The University of Queensland, Queensland, Australia) for the Cxcr5-Promoter-Thy1.1 reporter vector; NIH Tetramer Core Facility (NIH Contract 75N93020D00005 and RRID:SCR_026557) for providing I-A(b) Influenza A NP311–325 (QVYSLIRPNENPAHK) tetramer-APC; and Drs. Avinash Bhandoola (NIH/National Cancer Institute, Bethesda, MD, USA) and Jason G. Cyster (University of California, San Francisco, CA, USA) for discussion.

Funding includes NIH grants AI130232, AI167986, and AI177467 (to H. Hu); NIH training grants T32AI007051 and T32GM008361 (to A.R. Schroeder); and University of Alabama at Birmingham Center for AIDS Research Grant P30AI027767–26.

Footnotes

Disclosures: The authors declare no competing interests exist.

Data availability

ATAC-seq (GSE278494) data generated in this study have been deposited at GEO and are publicly available as of the date of publication. Accession numbers of subseries are listed in the Table S1. All the data described in the text are available in main figures and supplementary figures. Any additional information required to reanalyze the data reported in this paper is available from the corresponding author upon request.

References

  1. Afzali B, Grönholm J, Vandrovcova J, O’Brien C, Sun HW, Vanderleyden I, Davis FP, Khoder A, Zhang Y, Hegazy AN, et al. 2017. BACH2 immunodeficiency illustrates an association between super-enhancers and haploinsufficiency. Nat. Immunol. 18:813–823. 10.1038/ni.3753 [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Ballesteros-Tato A, León B, Graf BA, Moquin A, Adams PS, Lund FE, and Randall TD. 2012. Interleukin-2 inhibits germinal center formation by limiting T follicular helper cell differentiation. Immunity. 36:847–856. 10.1016/j.immuni.2012.02.012 [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Buenrostro JD, Wu B, Chang HY, and Greenleaf WJ. 2015. ATAC-seq: A method for assaying chromatin accessibility genome-wide. Curr. Protoc. Mol. Biol. 109:21.29.21–21.29.29. 10.1002/0471142727.mb2129s109 [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Crotty S 2011. Follicular helper CD4 T cells (TFH). Annu. Rev. Immunol. 29:621–663. 10.1146/annurev-immunol-031210-101400 [DOI] [PubMed] [Google Scholar]
  5. Crotty S 2014. T follicular helper cell differentiation, function, and roles in disease. Immunity. 41:529–542. 10.1016/j.immuni.2014.10.004 [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Cui Y, Riedlinger G, Miyoshi K, Tang W, Li C, Deng CX, Robinson GW, and Hennighausen L. 2004. Inactivation of Stat5 in mouse mammary epithelium during pregnancy reveals distinct functions in cell proliferation, survival, and differentiation. Mol. Cell Biol. 24:8037–8047. 10.1128/mcb.24.18.8037-8047.2004 [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Danecek P, Bonfield JK, Liddle J, Marshall J, Ohan V, Pollard MO, Whitwham A, Keane T, McCarthy SA, Davies RM, and Li H. 2021. Twelve years of SAMtools and BCFtools. Gigascience. 10:giab008. 10.1093/gigascience/giab008 [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. DiToro D, Winstead CJ, Pham D, Witte S, Andargachew R, Singer JR, Wilson CG, Zindl CL, Luther RJ, Silberger DJ, et al. 2018. Differential IL-2 expression defines developmental fates of follicular versus nonfollicular helper T cells. Science. 361:eaao2933. 10.1126/science.aao2933 [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Fazilleau N, McHeyzer-Williams LJ, Rosen H, and McHeyzer-Williams MG. 2009. The function of follicular helper T cells is regulated by the strength of T cell antigen receptor binding. Nat. Immunol. 10:375–384. 10.1038/ni.1704 [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Fooksman DR, Schwickert TA, Victora GD, Dustin ML, Nussenzweig MC, and Skokos D. 2010. Development and migration of plasma cells in the mouse lymph node. Immunity. 33:118–127. 10.1016/j.immuni.2010.06.015 [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Geng J, Wei H, Shi B, Wang YH, Greer BD, Pittman M, Smith E, Thomas PG, Kutsch O, and Hu H. 2019. Bach2 negatively regulates T follicular helper cell differentiation and is critical for CD4+ T cell memory. J. Immunol. 202:2991–2998. 10.4049/jimmunol.1801626 [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Grant FM, Yang J, Nasrallah R, Clarke J, Sadiyah F, Whiteside SK, Imianowski CJ, Kuo P, Vardaka P, Todorov T, et al. 2020. BACH2 drives quiescence and maintenance of resting Treg cells to promote homeostasis and cancer immunosuppression. J. Exp. Med. 217:e20190711. 10.1084/jem.20190711 [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Heinz S, Benner C, Spann N, Bertolino E, Lin YC, Laslo P, Cheng JX, Murre C, Singh H, and Glass CK. 2010. Simple combinations of lineage-determining transcription factors prime cis-regulatory elements required for macrophage and B cell identities. Mol. Cell. 38:576–589. 10.1016/j.molcel.2010.05.004 [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Hu H, Huston G, Duso D, Lepak N, Roman E, and Swain SL. 2001. CD4(+) T cell effectors can become memory cells with high efficiency and without further division. Nat. Immunol. 2:705–710. 10.1038/90643 [DOI] [PubMed] [Google Scholar]
  15. Hu H, Wang B, Borde M, Nardone J, Maika S, Allred L, Tucker PW, and Rao A. 2006. Foxp1 is an essential transcriptional regulator of B cell development. Nat. Immunol. 7:819–826. 10.1038/ni1358 [DOI] [PubMed] [Google Scholar]
  16. Igarashi K, Kurosaki T, and Roychoudhuri R. 2017. BACH transcription factors in innate and adaptive immunity. Nat. Rev. Immunol. 17:437–450. 10.1038/nri.2017.26 [DOI] [PubMed] [Google Scholar]
  17. Jenkins MK, and Moon JJ. 2012. The role of naive T cell precursor frequency and recruitment in dictating immune response magnitude. J. Immunol. 188:4135–4140. 10.4049/jimmunol.1102661 [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Johnston RJ, Choi YS, Diamond JA, Yang JA, and Crotty S. 2012. STAT5 is a potent negative regulator of TFH cell differentiation. J. Exp. Med. 209:243–250. 10.1084/jem.20111174 [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Johnston RJ, Poholek AC, DiToro D, Yusuf I, Eto D, Barnett B, Dent AL, Craft J, and Crotty S. 2009. Bcl6 and Blimp-1 are reciprocal and antagonistic regulators of T follicular helper cell differentiation. Science. 325:1006–1010. 10.1126/science.1175870 [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Lahmann A, Kuhrau J, Fuhrmann F, Heinrich F, Bauer L, Durek P, Mashreghi MF, and Hutloff A. 2019. Bach2 controls T follicular helper cells by direct repression of Bcl-6. J. Immunol. 202:2229–2239. 10.4049/jimmunol.1801400 [DOI] [PubMed] [Google Scholar]
  21. Langmead B, and Salzberg SL. 2012. Fast gapped-read alignment with Bowtie 2. Nat. Methods 9:357–359. 10.1038/nmeth.1923 [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Leong YA, Chen Y, Ong HS, Wu D, Man K, Deleage C, Minnich M, Meckiff BJ, Wei Y, Hou Z, et al. 2016. CXCR5(+) follicular cytotoxic T cells control viral infection in B cell follicles. Nat. Immunol. 17:1187–1196. 10.1038/ni.3543 [DOI] [PubMed] [Google Scholar]
  23. Li J, Lu E, Yi T, and Cyster JG. 2016. EBI2 augments Tfh cell fate by promoting interaction with IL-2-quenching dendritic cells. Nature. 533:110–114. 10.1038/nature17947 [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Liu Q, Zhu F, Liu X, Lu Y, Yao K, Tian N, Tong L, Figge DA, Wang X, Han Y, et al. 2022. Non-oxidative pentose phosphate pathway controls regulatory T cell function by integrating metabolism and epigenetics. Nat. Metab. 4:559–574. 10.1038/s42255-022-00575-z [DOI] [PubMed] [Google Scholar]
  25. Love MI, Huber W, and Anders S. 2014. Moderated estimation of fold change and dispersion for RNA-seq data with DESeq2. Genome Biol. 15:550. 10.1186/s13059-014-0550-8 [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Mackay LK, Minnich M, Kragten NAM, Liao Y, Nota B, Seillet C, Zaid A, Man K, Preston S, Freestone D, et al. 2016. Hobit and Blimp1 instruct a universal transcriptional program of tissue residency in lymphocytes. Science. 352:459–463. 10.1126/science.aad2035 [DOI] [PubMed] [Google Scholar]
  27. Malek TR, and Castro I. 2010. Interleukin-2 receptor signaling: At the interface between tolerance and immunity. Immunity. 33:153–165. 10.1016/j.immuni.2010.08.004 [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Moriggl R, Topham DJ, Teglund S, Sexl V, McKay C, Wang D, Hoffmeyer A, van Deursen J, Sangster MY, Bunting KD, et al. 1999. Stat5 is required for IL-2-induced cell cycle progression of peripheral T cells. Immunity. 10:249–259. 10.1016/s1074-7613(00)80025-4 [DOI] [PubMed] [Google Scholar]
  29. Nurieva RI, Chung Y, Martinez GJ, Yang XO, Tanaka S, Matskevitch TD, Wang YH, and Dong C. 2009. Bcl6 mediates the development of T follicular helper cells. Science. 325:1001–1005. 10.1126/science.1176676 [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Ochiai K, Katoh Y, Ikura T, Hoshikawa Y, Noda T, Karasuyama H, Tashiro S, Muto A, and Igarashi K. 2006. Plasmacytic transcription factor Blimp-1 is repressed by Bach2 in B cells. J. Biol. Chem. 281:38226–38234. 10.1074/jbc.m607592200 [DOI] [PubMed] [Google Scholar]
  31. Ochiai K, Muto A, Tanaka H, Takahashi S, and Igarashi K. 2008. Regulation of the plasma cell transcription factor Blimp-1 gene by Bach2 and Bcl6. Int. Immunol. 20:453–460. 10.1093/intimm/dxn005 [DOI] [PubMed] [Google Scholar]
  32. Pepper M, Pagán AJ, Igyarto BZ, Taylor JJ, and Jenkins MK. 2011. Opposing signals from the Bcl6 transcription factor and the interleukin-2 receptor generate T helper 1 central and effector memory cells. Immunity. 35:583–595. 10.1016/j.immuni.2011.09.009 [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. Quinlan AR, and Hall IM. 2010. BEDTools: A flexible suite of utilities for comparing genomic features. Bioinformatics. 26:841–842. 10.1093/bioinformatics/btq033 [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. Ramírez F, Ryan DP, Grüning B, Bhardwaj V, Kilpert F, Richter AS, Heyne S, Dündar F, and Manke T. 2016. deepTools2: a next generation web server for deep-sequencing data analysis. Nucleic Acids Res. 44:W160–W165. 10.1093/nar/gkw257 [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. Robinson JT, Thorvaldsdóttir H, Winckler W, Guttman M, Lander ES, Getz G, and Mesirov JP. 2011. Integrative genomics viewer. Nat. Biotechnol. 29:24–26. 10.1038/nbt.1754 [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. Schroeder AR, Xia X, Nguyen K, Zhu F, Geng J, Sialer DO, and Hu H. 2023. Cutting edge: Bach2 integrates cytokine signals to arbitrate differentiation decisions between T follicular helper and Th17 lineages. J. Immunol. 211:1756–1761. 10.4049/jimmunol.2300354 [DOI] [PubMed] [Google Scholar]
  37. Schroeder AR, Zhu F, and Hu H. 2021. Stepwise Tfh cell differentiation revisited: New advances and long-standing questions. Fac. Rev. 10:3. 10.12703/r/10-3 [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Sidwell T, Liao Y, Garnham AL, Vasanthakumar A, Gloury R, Blume J, Teh PP, Chisanga D, Thelemann C, de Labastida Rivera F, et al. 2020. Attenuation of TCR-induced transcription by Bach2 controls regulatory T cell differentiation and homeostasis. Nat. Commun. 11:252. 10.1038/s41467-019-14112-2 [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. Tanaka H, Muto A, Shima H, Katoh Y, Sax N, Tajima S, Brydun A, Ikura T, Yoshizawa N, Masai H, et al. 2016. Epigenetic regulation of the Blimp-1 gene (Prdm1) in B cells involves Bach2 and histone deacetylase 3. J. Biol. Chem. 291:6316–6330. 10.1074/jbc.m116.713842 [DOI] [PMC free article] [PubMed] [Google Scholar]
  40. Tsukumo S. i., Unno M, Muto A, Takeuchi A, Kometani K, Kurosaki T, Igarashi K, and Saito T. 2013. Bach2 maintains T cells in a naive state by suppressing effector memory-related genes. Proc. Natl. Acad. Sci. USA. 110:10735–10740. 10.1073/pnas.1306691110 [DOI] [PMC free article] [PubMed] [Google Scholar]
  41. Tubo NJ, Pagán AJ, Taylor JJ, Nelson RW, Linehan JL, Ertelt JM, Huseby ES, Way SS, and Jenkins MK. 2013. Single naive CD4+ T cells from a diverse repertoire produce different effector cell types during infection. Cell. 153:785–796. 10.1016/j.cell.2013.04.007 [DOI] [PMC free article] [PubMed] [Google Scholar]
  42. Victora GD, and Nussenzweig MC. 2012. Germinal centers. Annu. Rev. Immunol. 30:429–457. 10.1146/annurev-immunol-020711-075032 [DOI] [PubMed] [Google Scholar]
  43. Villarino AV, Laurence AD, Davis FP, Nivelo L, Brooks SR, Sun HW, Jiang K, Afzali B, Frasca D, Hennighausen L, et al. 2022. A central role for STAT5 in the transcriptional programing of T helper cell metabolism. Sci. Immunol. 7:eabl9467. 10.1126/sciimmunol.abl9467 [DOI] [PMC free article] [PubMed] [Google Scholar]
  44. Vinuesa CG, Tangye SG, Moser B, and Mackay CR. 2005. Follicular B helper T cells in antibody responses and autoimmunity. Nat. Rev. Immunol. 5:853–865. 10.1038/nri1714 [DOI] [PubMed] [Google Scholar]
  45. Wang H, Geng J, Wen X, Bi E, Kossenkov AV, Wolf AI, Tas J, Choi YS, Takata H, Day TJ, et al. 2014. The transcription factor Foxp1 is a critical negative regulator of the differentiation of follicular helper T cells. Nat. Immunol. 15:667–675. 10.1038/ni.2890 [DOI] [PMC free article] [PubMed] [Google Scholar]
  46. Yao Z, Cui Y, Watford WT, Bream JH, Yamaoka K, Hissong BD, Li D, Durum SK, Jiang Q, Bhandoola A, et al. 2006. Stat5a/b are essential for normal lymphoid development and differentiation. Proc. Natl. Acad. Sci. USA. 103:1000–1005. 10.1073/pnas.0507350103 [DOI] [PMC free article] [PubMed] [Google Scholar]
  47. Yu D, Rao S, Tsai LM, Lee SK, He Y, Sutcliffe EL, Srivastava M, Linterman M, Zheng L, Simpson N, et al. 2009. The transcriptional repressor Bcl-6 directs T follicular helper cell lineage commitment. Immunity. 31:457–468. 10.1016/j.immuni.2009.07.002 [DOI] [PubMed] [Google Scholar]
  48. Zhang H, Dai D, Hu Q, Yang F, Xue Y, Li F, Shen N, Zhang M, and Huang C. 2021. Bach2 attenuates IL-2R signaling to control Treg homeostasis and Tfr development. Cell Rep. 35:109096. 10.1016/j.celrep.2021.109096 [DOI] [PubMed] [Google Scholar]
  49. Zhang H, Hu Q, Zhang M, Yang F, Peng C, Zhang Z, and Huang C. 2019. Bach2 deficiency leads to spontaneous expansion of IL-4-producing T follicular helper cells and autoimmunity. Front Immunol. 10:2050. 10.3389/fimmu.2019.02050 [DOI] [PMC free article] [PubMed] [Google Scholar]
  50. Zhu F, McMonigle RJ, Schroeder AR, Xia X, Figge D, Greer BD, González-Avalos E, Sialer DO, Wang YH, Chandler KM, et al. 2023. Spatiotemporal resolution of germinal center Tfh cell differentiation and divergence from central memory CD4+ T cell fate. Nat. Commun. 14:3611. 10.1038/s41467-023-39299-3 [DOI] [PMC free article] [PubMed] [Google Scholar]
  51. Zhu J, Yamane H, and Paul WE. 2010. Differentiation of effector CD4 T cell populations (*). Annu. Rev. Immunol. 28:445–489. 10.1146/annurev-immunol-030409-101212 [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

supplementary Table S1
NIHMS2206137-supplement-supplementary_Table_S1.pdf (284.2KB, pdf)
SourceDataF2
NIHMS2206137-supplement-SourceDataF2.pdf (5.4MB, pdf)
SouceDataFS2
NIHMS2206137-supplement-SouceDataFS2.pdf (5.3MB, pdf)
1
NIHMS2206137-supplement-1.pdf (1.1MB, pdf)

Data Availability Statement

ATAC-seq (GSE278494) data generated in this study have been deposited at GEO and are publicly available as of the date of publication. Accession numbers of subseries are listed in the Table S1. All the data described in the text are available in main figures and supplementary figures. Any additional information required to reanalyze the data reported in this paper is available from the corresponding author upon request.

RESOURCES